Progress in Oceanography 133 (2015) 43–54

Contents lists available at ScienceDirect

Progress in Oceanography

journal homepage: www.elsevier .com/ locate /pocean

Optical techniques for remote and in-situ characterization of particlespertinent to GEOTRACES

http://dx.doi.org/10.1016/j.pocean.2014.09.0070079-6611/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail addresses: emmanuel.boss@maine.edu (E. Boss), lguidi@obs-vlfr.fr

(L. Guidi), mrichardson@ocean.tamu.edu (M.J. Richardson), stemmann@obs-vlfr.fr(L. Stemmann), wgardner@ocean.tamu.edu (W.D. Gardner), jkbishop@berkeley.edu(J.K.B. Bishop), boba@ldeo.columbia.edu (R.F. Anderson), sherrell@marine.rutgers.edu (R.M. Sherrell).

Emmanuel Boss a,⇑, Lionel Guidi b, Mary Jo Richardson c, Lars Stemmann d, Wilford Gardner c,James K.B. Bishop e, Robert F. Anderson f, Robert M. Sherrell g

a School of Marine Sciences, 5706 Aubert Hall, University of Maine, Orono, ME 04473, USAb Laboratoire d’Oceanographie de Villefranche (CNRS), Observatoire Océanographique de Villefranche, 181, Chemin du Lazaret, 06234 Villefranche sur mer, Francec Texas A&M University, 3146 TAMU, College Station, TX 77843, USAd Sorbonne Universités (Université Pierre et Marie Curie, Paris 6), Observatoire Océanologique de Villefranche-sur-Mer, 06230 Villefranche-sur-Mer, Francee University of California Berkley, 499 McCone Hall, Berkeley, CA 94720, USAf Lamont-Doherty Earth Observatory of Columbia University, 231 Comer, 61 Route 9W, PO Box 1000, Palisades, NY 10964, USAg Institute of Marine and Coastal Sciences, Dept. of Earth and Planetary Sciences, Rutgers University, 71 Dudley Rd., New Brunswick, NJ 08901, USA

a r t i c l e i n f o

Article history:Available online 14 October 2014

a b s t r a c t

Field and laboratory characterization of marine particles is laborious and expensive. Proxies of particleproperties have been developed that allow researchers to obtain high frequency distributions of suchproperties in space or time. We focus on optical techniques used to characterize marine particles in-situ,with a focus on GEOTRACES-relevant properties, such as bulk properties including particle mass, cross-sectional area, particle size distribution, particle shape information, and also single particle optical prop-erties, such as individual particle type and size. We also address the use of optical properties of particlesto infer particulate organic or inorganic carbon. In addition to optical sensors we review advances inimaging technology and its use to study marine particles in situ. This review addresses commerciallyavailable technology and techniques that can be used as a proxy for particle properties and the associateduncertainties with particular focus to open ocean environments, the focus of GEOTRACES.

� 2014 Elsevier Ltd. All rights reserved.

Introduction to the role of particles in GEOTRACES

Riverine particles and atmospheric aerosols serve as input oftrace elements to seawater. Trace elements are removed from sea-water by active uptake by phytoplankton, subsequent grazing andsedimentation as well as via adsorption to particles. They are alsointernally recycled within the ocean through physical and biogeo-chemical processes (for recent summaries, see Anderson andHayes, 2014; Jeandel et al., 2015). Consequently, an accurateknowledge of the abundance, distribution and composition of par-ticles in the ocean is essential to the mission of the GEOTRACESprogram, an international study of the marine biogeochemicalcycles of trace elements and their isotopes <www.geotraces.org>.Collection of particles in discrete samples, followed by the chemi-cal characterization of particle composition, is widely used inGEOTRACES to define the role of particles in trace element cycles(Lam et al., 2015; McDonnell et al., 2015). The inherent spatial

and temporal limitations of discrete sampling potentially intro-duce a bias into the interpretation of the results due to an incom-plete description of the particle field.

Optical techniques allow for sampling at much greater spatialand temporal resolution, providing a more informed context forinterpreting the results from discrete samples as documented below(Fig. 1). In addition, preliminary application of optical techniquesduring early GEOTRACES cruises has helped to establish an impor-tant role for intermediate and benthic nepheloid layers particles inthe removal of radionuclides and, presumably, of other trace ele-ments from seawater (Deng et al., 2014; Hayes et al., in press). Infor-mation presented in this review will facilitate the incorporation ofadditional sophisticated optical techniques in future GEOTRACESexpeditions. This information will also be of value in planning otherprograms, such as those studying the marine biogeochemical cyclesof carbon, of major nutrients, and of anthropogenic contaminants.

Optical properties and their sensitivity to mass, size,composition, shape, and packing of marine particles

Inherent optical properties (IOP) of seawater and its constitu-ents are independent of sun illumination and can be measured in

Fig. 1. Example of a section of beam attenuation coefficient due to particles (cp)collected during the South Atlantic Ventilation Experiment (SAVE) in February,1989. Shows location of section (red) from North to South. A 25-cm path-lengthSeaTech transmissometer (acceptance angle 1.03�). Note bottom expansion of thedepth scale for higher resolution. Concentrations of resuspended boundary sedi-ments exceeded 1500 lg l�1 in the Argentine Basin (Wood, 1993; Gardner et al.,2014) but were seldom greater than 100 lg l�1 in the Brazil basin (Richardson et al.,1990). (For interpretation of the references to color in this figure legend, the readeris referred to the web version of this article.)

44 E. Boss et al. / Progress in Oceanography 133 (2015) 43–54

the laboratory and the field. Absorption and angular scattering arethe processes through which monochromatic (i.e. narrow wave-length band) and collimated (i.e. highly directional) light from asource is lost out of a light beam through annihilation or redirec-tion, respectively. The attenuation coefficient (c) is the sum ofthe absorption coefficient (a) and the scattering coefficient (b),which is the integral of all directions of angular scattering. An addi-tional IOP is inelastic scattering; the process whereby lightabsorbed at one wavelength is emitted at another wavelength.

To first order, optical properties of particles vary with concen-tration. To second order, they vary with size, shape, internal struc-ture and packing. While the interaction of matter with light iscomplicated (e.g. Jonasz and Fournier, 2007, and Sosik, 2008, forrecent reviews), some generalizations can be made, in particularwith respect to the effect of particle size. For particles much smal-ler than the wavelength, scattering is fore-aft symmetric anddecreases as wavelength to the negative 4th power, whereasabsorption for such particles is proportional to their volume. Forparticles much larger than the wavelength, absorption, scatteringand attenuation correlate with cross-sectional area (i.e. their sha-dow, e.g. Jonasz and Fournier, 2007), exhibiting relatively littlespectral variations in scattering, and hence are relatively insensi-tive to composition. To first order, the mass normalized opticalproperty of a single particle can be predicted by the ratio of its size(e.g. the diameter of a sphere with a similar cross-section) to thewavelength of light times the difference between its index ofrefraction and that of water (e.g. Figs. 3–6 in Stemmann andBoss, 2012). While mass normalized absorption decreases approx-imately monotonically with particle size (Duysens, 1956), massnormalized scattering exhibits a resonance behavior as a functionof size. Scattering reaches a maximum for particles of a fewmicrons, then decreases as diameter�1 for larger solid particles(Baker et al., 2001; Fig. 2).

Because particulate matter (PM) concentrations within thewater column change over four orders of magnitude (e.g. PM vary-ing from 5 to >12,000 lg kg�1 as a function of depth and location inthe open ocean, Brewer et al., 1976; Gardner et al., 1985), the pri-mary variability in optical properties in the ocean (from the seawa-ter and the particles) is due to particle concentration. Additionally,optical properties are sensitive to particle size, shape and compo-sition. Angular scattering for particles is a strong function of parti-cle size; in general, the larger the particle, the more the energy is

Fig. 2. Mass-specific side-scattering (reported as turbidity) values (Kbs) as afunction of size for plastic beads (triangles. Solid triangles is calculated for densityof 2.65 g cm�3) and natural particles (solid circles) (from Baker et al., 2001). Notethe resonance near 1micron and the decrease as diameter�1 for larger sizes.

scattered in the same direction the light is propagating. Composi-tion effects are mediated through the index of refraction of thematerial particles that are composed of both organic (lipids, sugar,proteins, starch) and inorganic materials (opal, calcite, clay miner-als). These materials have indices of refraction that are significantlydifferent from water (n > 1.12, Aas, 1996, where n is measured rel-ative to water), and hence cause the particles to interact signifi-cantly with light (the more different the index of refraction of aparticle relative to water, the more intense the scattering from thatparticle). Because phytoplankton and other organic particles (suchas bacteria) often have a large fraction of their volume occupied byfluid, their ‘effective’ index of refraction (that is if we model themas homogeneous particles) is quite similar to that of water(n � 1.05, Aas, 1996).

For large populations of particles, composition was found to pri-marily modulate the ratio of backward to forward scattered light(e.g. Twardowski et al., 2001; Fournier et al., 2014). When the opti-cal properties such as attenuation and scattering are normalized bydry-mass (that is without the interstitial water, as when obtainingdry weight PM), the mass specific optical property varies by only afactor of two between organic and inorganic particles (Babin et al.,2003).

Aggregation of oceanic particles most often results in particleshaving a large fluid fraction and significant amount of mass(Alldredge and Jackson, 1995), and so will change angular scatter-ing while nearly maintaining mass normalized optical properties(Boss et al., 2009a; Slade et al., 2011). It has been observed thatthe bulk optical properties of particle populations are propor-tional to the total cross-sectional area of the disaggregated parti-cles (e.g. as measured with Coulter counters, see for exampleFig. A1, in Behrenfeld and Boss, 2006). Hence, aggregation anddisaggregation of oceanic particles tend to have a relatively smalleffect on mass normalized optical properties, as the cross-sectional area of the solid material is nearly conserved. Howeverrare and large heterogeneous aggregates are not well sampled bybulk optical methods. For these aggregates, imaging methods arebetter (see below; McDonnell et al., 2015). Similar arguments asfor heterogeneous aggregates may apply to transparent exo-poly-mers (TEP, whose primary material is made from high index ofrefraction building block, but with a lot of added water,Alldredge et al., 1993).

E. Boss et al. / Progress in Oceanography 133 (2015) 43–54 45

Commercial technology to measure optical properties in-situ

Commercially available in-situ optical instruments can be usedto obtain estimates of the beam-attenuation, backscattering, side-scattering and absorption coefficients, which are then used asproxies for particle concentration. A list of worldwide datasetswith such measurements is provided in Appendix A. The reasonwe use the word estimate is that no instrument measures exactlythe property it is supposed to measure. Attenuation meters have afinite acceptance angle, which results in a de facto upper size cut-off sensitivity to particles (Boss et al., 2009b), and, when pathlength is short and/or particle concentrations are large, may bebiased due to multiple scattering (e.g. Leymarie et al., 2010). Back-scattering sensors measure light scattered at a single angle in theback direction. This single-angle measurement is used to estimatethe full back-hemispherical scattering. This conversion has anassociated uncertainty around 10% for typical marine particles(Boss et al., 2001a; Sullivan et al., 2013), and in very turbid condi-tions needs to be corrected for attenuation along the path. Theexact angular response of side-scattering sensors is most oftenunknown and hence output is not in strictly physical units; ratherit is calibrated with a turbidity standard. Finally, the vetted com-mercial instruments available for measuring the absorption coeffi-cient in-situ (WETLabs ac-9 & ac-S) have a deficiency in collectingall the scattered light and hence needs to be corrected in a proce-dure for which there is still no consensus and that introducesuncertainties on the order of a few tens of percent in the blue por-tion of the spectrum (e.g. Leymarie et al., 2010; Röttgers et al.,2013). Note, however, that a novel sensor has recently becomecommercial which should, in principle, provide in-situ absorptionthat does not need to be corrected for scattering (Turner Design’sICAM based on Fry et al., 1992). Given the above, comparing mea-surements by different sensors requires that instrument specifica-tions are known and comparable.

Use of single optical properties to obtain particle mass, size andcomposition

The introduction of the first commercial transmissometersspawned the investigation of PM and particulate organic carbonconcentration (POC) relationships with beam attenuation coeffi-cient (Bishop, 1986, 1999; Bishop et al., 1999, 1992; Bishop andJoyce, 1986; Bishop and Wood, 2008; Cetinic et al., 2012;Gardner, 1989a; Gardner et al., 1985, 2003, 1993; Hill et al.,2011; Karageorgis et al., 2008, 2014; Neukermans et al., 2012).The beam attenuation coefficient was found to be a better proxyfor POC than for PM for waters dominated by biogenic particles,having a smaller intercept and a significantly higher correlationcoefficient (Bishop, 1999; Gardner et al., 2003).

Remote sensing requirements and the need to understandcoastal and near bottom processes, stimulated the developmentof PM/POC – backscattering proxies (Boss et al., 2009c; Cetinicet al., 2012; Neukermans et al., 2012; Stramski et al., 1999). Theavailability of relatively cheap and large-depth capable side-scat-tering (turbidity) sensors encouraged studies relating side-scatter-ing and POC or PM (Baker et al., 2001; Bishop and Wood, 2008;Boss et al., 2009c; Bishop et al., 2012; Cetinic et al., 2012).

Gardner et al. (1993) found in a North Atlantic study, that inclu-sion of the particles settling to the bottom of a Niskin bottle(Gardner, 1977) made no significant improvement to the correla-tion between PM and cp; it did, however, increase the slope ofthe regression by 40%, demonstrating that the settled particles(likely large aggregates) contributed significantly to cp. Bishopand co-workers (Bishop, 1999; Bishop et al., 1999; Bishop andWood, 2008, and Bishop et al., 2012) have found that inclusion of

the >51 lm particle size-fraction improves the correlation of POCto cp relationships. They found that a single scaling factor relatesPOC and cp in oligotrophic waters near Hawaii and in productivewaters of the Oyashio (see Fig. 3AB). In contrast, they found side-scattering data to severely underestimate euphotic zone POC con-centrations (Fig. 3CD). Cetinic et al. (2012), using data from theNorth Atlantic, have found that POC to cp relationships hold equallywell above and below the euphotic layer. In contrast, they foundthat backscattering/cp and side-scattering/cp ratio varied withdepth.

The variability in mass specific optical properties, when com-pared across studies, has been found to be constrained within±50% despite some variability in the analytical methods used toobtain mass (Boss et al., 2009c; Cetinic et al., 2012; Hill et al.,2011; Neukermans et al., 2012) and large variability in theenvironments sampled. Note that in all these studies measurementshave been done using red or near-infra-red wavelengths(650 nm < k < 880 nm) to minimize the effect of dissolved substanceabsorption on the attenuation coefficient, as well as to minimize theeffect of particulate absorption on the scattering coefficient.

Shifrin (1988) has introduced the ‘method of fluctuation’whereby fluctuation in beam attenuation can be used to providean average particle size estimate. Briggs et al. (2013) have devel-oped a similar method (though derived differently) to obtain anaverage particle size from backscattering and beam attenuationand have validated it in the laboratory against an establishedmethod (laser diffraction). The isolation of spikes in beam attenu-ation coefficient and scattering signals has been shown to be anindex of both zooplankton and aggregate particles (Bishop andWood, 2008; Gardner et al., 2000).

As noted above, there are locations where local or advectedresuspended sediments increase particle concentrations and theassociated cp. These occurrences can be understood with the addi-tional measurement of the backscattering ratio (see below). Duringthe GEOTRACES program, evidence of active resuspension anddeposition has been found to be important in scavenging someparticle-reactive radionuclides in bottom waters (Hayes et al., inpress; Anderson and Hayes, 2015) and is likely to have other bio-geochemical impacts waiting to be discovered. Global mapping ofareas of resuspension is needed for determining likely sites forfurther exploration (e.g., Gardner et al., 2014, Fig. 1).

Combination of several optical properties

Spectral optical propertiesSpectral beam attenuation of particles (the difference between

measurements of unfiltered samples minus those filtered by anin-situ 0.2 lm filter) has been shown to provide a robust size indexof particles in the size range of 0.2–20 lm (Boss et al., 2001a; Sladeet al., 2011). A relationship between the particle beam attenuationspectral-slope and a power-law particulate size distribution (PSD)is predicted by idealized theory (e.g. Mie theory which assumesparticles are spheres and of homogeneous optical properties, e.g.Boss et al., 2001b). A commercial switching device (FlowControl)to automate the filtered/non-filter technique for in-line or in-situmeasurements of particulate properties has recently been com-mercialized (Sequoia Scientific). Spectral absorption of particleshas been measured in situ using the same filtration system at bothlow and high spectral resolution. Low spectral resolution data hasbeen used to estimate in-situ chlorophyll concentration from thepeak height of the 676 nm Sorret band in an oligotrophic lake(Davis et al., 1997; Boss et al., 2007). High spectral resolution par-ticulate absorption data were shown to provide concentrations ofin-situ phytoplankton pigments and hence to extract informationon the underlying phytoplankton population (Moisan et al., 2011;Chase et al., 2013).

0 0.2 0.4 0.6 0.8 1 1.2 1.4

ALOHAM01 dayM02 ntM03 dayM04 nt

0 1 2 3 4 5

0 1 2 3 4 5

1000

900

800

700

600

500

400

300

200

100

0

Dep

th (

m)

K2(days 211-218)

K2(days 221-228)

M07 dayM08 nt

M09 dayM10 nt

(A) (B)

(D)(C)

CTD dawn

CTD dawnCTD dusk

CTD duskCTD duskCTD dawn

0 0.2 0.4 0.6 0.8 1 1.2 1.4

1000

900

800

700

600

500

400

300

200

100

0

Dep

th (

m)

Particulate Organic Carbon (µM) Particulate Organic Carbon (µM)

POCscat

POCCp POCCp

POCscat

Fig. 3. Distribution of POC (in lM) from Multiple Unit Large Volume in-situ Filtration System (MULVFS) at oligotrophic station ALOHA and biologically productive station K2in the Pacific Ocean overlain on profiles of the scaled beam attenuation coefficient, and scaled side-scattering. (A and B) Dawn/Dusk POCCp and POC as function of depth withPOC at ALOHA and K2. (C and D) Seapoint turbidity sensor scaled to match POC at 310 m. POCCp estimates of POC are better than POCscat. Zooplankton captured by MULVFScontributed an estimated 0.9 lM (night) and 0.2 lM (day) POC for K2 at 10 m (other samples were not significantly affected). Figure modified from Bishop and Wood (2008).

46 E. Boss et al. / Progress in Oceanography 133 (2015) 43–54

Most recently, a method that estimates the iron content of lith-ogenic particles using their absorption at blue wavelengths hasbeen described (Estapa et al., 2012). Such a method, if proved towork with in-situ sensors, could pave the way for measurementof in-situ particle bound iron.

Since the 1960s fluorescence measurements have been used reg-ularly to sample phytoplankton pigments in situ. This method isused to quantify pigments in situ and to differentiate phytoplanktonfunctional types. Interpretation of the signals is not trivial particu-larly if quantitative results are sought (see recent review byMcIntyre et al., 2010).

Multiple angle scatteringDiffraction-based techniques have been used for fifteen years to

obtain size resolved information of PSD in situ (that is, which min-imally affect aggregation state) by inversion of near-forward scat-tering measurements made using the Laser In Situ Scattering andTransmissometry (LISST) family of in situ sensors (Agrawal andPottsmith, 2000). Sizes resolved with different LISST instrumentsspan from about 2 to 500 lm. The inversion is most sensitive to aparticle’s cross sectional area. Hence, non-sphericity of marine par-ticles results in smearing across size classes as different orientationsare presented to the sensor (Karp-Boss et al., 2007) while providinga useful measure of the particulate cross-sectional area (which isproportional to the surface area for randomly oriented convexparticles).

The LISST sensors were primarily designed to work in turbidenvironments (they have a 5-cm pathlength) and hence have a rel-atively low signal-to-noise ratio in clear water. Nonetheless several

studies have used them in the upper ocean with success(Neukermans et al., 2012; Reynolds et al., 2010).

Backscattering ratioThe ratio of particle backscattering to the particle scattering

coefficient, the backscattering ratio, has been found to be a usefuldescriptor of the bulk composition of the particle assemblage. Itvaries in the field by about a factor of 6 from 0.5% associated withphytoplankton to about 3% associated with highly refractive litho-genic particles (e.g. Boss et al., 2009c; Loisel et al., 2007;Twardowski et al., 2001). An extreme value of 5% has also beenreported (McKee et al., 2009).

Optical sensing methods to obtain information on particulateinorganic carbon concentration

Cross-polarized transmission

Birefringence refers to the ability of a mineral crystal to split anincident beam of linearly polarized light into two beams of unequalvelocities (corresponding to two different refractive indices of thecrystal), which subsequently recombine to form a beam of lightthat is no longer linearly polarized.

The extreme birefringence of CaCO3 makes it light up whenthe sample is held between crossed polarizers and viewed usingtransmitted light. This characteristic mineralogical property ofCaCO3 is widely used as a means of identification. Calcium car-bonate particles have an oceanic concentration range of 0.005–40 lM. It has been demonstrated that a bench top spectropho-

E. Boss et al. / Progress in Oceanography 133 (2015) 43–54 47

tometer can be used to detect particulate inorganic carbon (PIC,Guay and Bishop, 2002; IOCCG, 2011). Experiments with varyingcoccolith concentration in suspension with and without varyingconcentrations of non-birefringent material have established alinear concentration response to PIC. The measurement of bire-fringent photons requires simultaneous measurement of trans-mission to correct for scattering losses, although these lossesare minor in most cases. Birefringence is measured as: B = Bobs/T0.5, where B = corrected birefringence, Bobs = raw birefringencesignal and T (as a fraction) is the measured fractional transmis-sion (relative to particle free water) at the same wavelength. Alow power (<0.5 W) PIC sensor has been used in diverse oceanenvironments ranging from polar regions to the equator, includ-ing oligotrophic to coastal and oxic to sub-anoxic environments(Bishop, 2009). The sensor has been under development (withWETLabs) since 2010 with deployments on CTDs in the subarcticN Pacific in 2012 and 2013. Repeat profiles at station PAPA showa detection limit of 0.01 lM and deepwater reproducibility is0.02 lM. Several mechanical/configuration issues have been iden-tified that need to be addressed before the sensor is availablecommercially.

Acid-labile method

‘‘Acid labile backscattering’’ is calculated as the differencebetween total and acidified backscattering (measured after add-ing a weak acid in order to drop the pH of the seawater/particlesuspension below the dissociation point for calcite, Balch et al.,2001). This setup is ideal for semi-continuous shipboard mea-surements over a variety of spatial and temporal scales. Typically,acid-labile backscattering measurements can be made once everyfew minutes aboard ship, allowing spatial resolution of �1 km attypical ship speeds (e.g. Balch and Utgoff, 2009). Conveniently(and fortuitously), due to the typical coccolith backscatteringcross section, the magnitude of the acid-labile backscattering(in units m�1) is approximately numerically equal to the PIC con-centration (in moles m�3, Op. Cit.).

Remotely-sensed ocean color

Algorithms to obtain particle properties from ocean colorremote sensing come in several varieties. Correlation-based algo-rithms are used to map chlorophyll (O’Reilly et al., 1998), POC(Son et al., 2009; Stramski et al., 1999, 2008), PM (Vantrepotteet al., 2011), PIC (Balch et al., 2007), and phytoplankton functionalgroups (Alvain et al., 2005; IOCCG, 2014). They tend to work best inopen ocean conditions where the assumptions that phytoplanktonand associated constituents co-vary and primarily modulate oceancolor are valid. Semi-analytical algorithms have been developedwhich, in addition to empirical input, use radiative-transfer theoryto invert optical constituents in the open ocean. These algorithms,most often, obtain the combined absorption of non-phytoplanktonand dissolved material, phytoplankton absorption, the associatedchlorophyll concentration, and the backscattering coefficient (seeIOCCG, 2006; Werdell et al., 2013). Semi-analytical algorithms pro-vide information regarding size (Loisel et al., 2006; Kostadinovet al., 2009; Berwin et al., 2011) and phytoplankton compositionas well (Kostadinov et al., 2010).

A resource with examples using ocean-color remote sensing fora variety of ocean related studies with many case studies is pro-vided at: http://www.ioccg.org/handbook.html. One is cautionedthat many ocean-color based algorithms provide output withoutassociated uncertainties. In addition, empirical algorithms that

have been tuned with specific data obtained in a specific regionat a given time may not be applicable to other conditions.

Optical imaging of particle abundance and size distribution

Relevant to the GEOTRACES objectives, imaging systems pro-vide particle size spectra and characterize individual particles athigh frequency. These are useful to estimate (1) macrozooplanktonabundance (Checkley et al., 2008; Stemmann et al., 2008a), (2) par-ticle flux through the water column (Guidi et al., 2008; Guidi et al.,2009; Jackson and Burd, 2002; Jouandet et al., 2011), and (3) par-ticle size distribution in mesoscale spatial patterns (Guidi et al.,2012; see e.g. Fig. 4). They can be used to test the influence of par-ticle size distribution on carbon and trace element (TE) fluxes(Bishop, 2009; Burd et al., 2007; Stemmann et al., 2004a;Stemmann et al., 2004b), and identify depths where aggregationtakes place that may affect the distribution of elements (Burdand Jackson, 2009).

Optical imaging systems such as digital cameras have been usedsince the 1980s to study in-situ concentrations of particles largerthan tens of micron including marine snow with an equivalentspherical diameter (ESD) >500 lm (Alldredge, 1998; Davis et al.,2005; Goldthwait and Alldredge, 2006; Gorsky et al., 1992; Honjoet al., 1984; McDonnell et al., 2015; Walsh and Gardner, 1992).Since then, the technology has evolved allowing fast laboratory-based or in-situ measurements of particle size distribution froma few microns to a few millimetres. The particle size range ‘‘seen’’by these instruments is limited by the resolution of the imagingsensor and the volume of water sampled.

These instruments have been used to study biogeochemicalcycles combining particle information from camera systems andsediment trap measurements of particle flux in order to betterestimate particle sinking rate, a characteristic influencing particleexport (Asper, 1987; Asper et al., 1992; Diercks and Asper, 1997;Gardner and Walsh, 1990; Guidi et al., 2008; Iversen et al., 2010;Jackson et al., 1997; Jouandet et al., 2011; McDonnell andBuesseler, 2010; Peterson et al., 2005; Walsh and Hunter,1992). Systems such as the In Situ Ichthyoplankton Imaging Sys-tem ISIIS (Cowen and Guigand, 2008), the Zooplankton Visualiza-tion and Imaging System ZOOVIS (Benfield et al., 2007) and theShadowed Image Particle Profiler and Evaluation Recorder (SIP-PER, Samson et al., 2001) allow counting and sizing of large zoo-plankton from images. The Video Plankton Recorder (VPR, Daviset al., 2005) and the Underwater Vision Profiler (UVP, Picheralet al., 2010) provide high temporal resolution particle imagingand size individual particles >�100 lm. However the larger watervolume per image sampled by the UVP (Volume UVP � 100 timesgreater than the VPR) allows better estimation with higher verti-cal resolution of concentrations of rare large particles (>1 mm).Several systems utilizing holographic imaging such as the LISST-holo (Braithwaite et al. 2012) have also been developed for thepurpose of imaging individual particles with sizes >�50 lmin situ (Hobson et al., 1997; Katz et al., 1999; Alexander et al.,2000; Pfitsch et al., 2007). Other optical instruments such asthe laser optical plankton counter (LOPC) can be used to quantifyzooplankton and particles based on the opacity and size of parti-cles (Checkley et al., 2008; Gonzalez-Quiros and Checkley, 2006;Jackson and Checkley, 2011). Validation is critical to insure parti-cles are well identified.

While most of the instruments presented above allow imagingand measurement of large (>100s lm) particles there is increasinginterest in quantifying nano- and microplankton particles. Labora-tory, on-vessel or in-situ instruments such as the FlowCam (FlowImaging), the Imaging FlowCytobot (McLane), or CytoSense

Fig. 4. Example of data obtained from a deployment of the UVP through the Oxygen Minimum Zone of the Northern Indian Ocean in March 2009 (A). Particulate carbon fluxprofile (red line) estimated from particle size distribution and concentration according to Guidi et al. (2008). (B) Particle abundance (PSD) for particles with equivalentspherical diameter (ESD, determined from particles cross-section) varying from 100 lm to 2 cm, binned every 200 m down to 1200 m compared to a theoretical PSD withslope equal to �4 in red. (C) Difference between the theoretical and the observed PSD with examples of aggregates from the UVP images found where residuals weremaximum. In this example, large aggregate abundance increases with depth below 150 m, presumably reflecting formation and loss of aggregates sinking down to the deepsea (Roullier et al., 2014). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

48 E. Boss et al. / Progress in Oceanography 133 (2015) 43–54

(CytoBuoy) allow this based on imaging and/or flow cytometery(Dubelaar and Gerritzen, 2000; Olson and Sosik, 2007; Sierackiet al., 1998).

Deployment strategies

Sensors to measure marine particles have been routinelydeployed on CTD rosettes (Bishop and Joyce, 1986; Bishop et al.,1992; Bishop et al., 2012; Bishop, 1999; Bishop and Wood, 2008;Dickey et al., 2006; Gardner et al., 1985; Gardner et al., 1993;Gardner et al., 2001; Hill et al., 2011; Karp-Boss et al., 2004;Stemmann et al., 2008a), during deployments of large volume in-situ filtration systems (Bishop, 1999; Bishop et al., 1999; Bishopet al., 2012; Bishop and Wood, 2008; Moran et al., 1999), onslow-drop packages (packages that fall under their own weight,Barnard et al., 1998), towed pumping systems, (Holser et al.,2011), AUVs and profiling floats (Bishop et al., 2002; Bishop andWood, 2009; Bishop et al., 2004; Boss et al., 2008; Checkley et al.,2008; Dickey et al., 2008), submarines, bottom tripods (Gardneret al., 1985; Slade et al., 2011) and moorings (Dickey et al., 2006;Gardner, 1989a, 1989b; Richardson et al., 1993). Consideration ofdeployment platform should include potential of platform interfer-ence with particles (e.g. by inducing turbulence that break aggre-gates) and the ability of the platform to deal with the weight,size, power and data streaming demands of a particular sensor.

Optical technology for measuring particles and their properties thatcould enhance GEOTRACES cruises

Recommendations for sensors and sensor systems that could beincluded in GEOTRACES cruises to provide information regardingparticles are provided in Table 1, with relationships between cer-tain bulk particle properties and optical variables are provided inTable 2.

The US GEOTRACES intercalibration experiment cruises haveprovided a test bed for transmissometer scattering, and PIC sensordeployments during trace metal rosette casts and in-situ pumpcasts (e.g. Appendix B).

Calibration strategies and resolution issues in nearly particlefree oceanic waters

Calibration and clean techniques with particle sampling andoptical measurements are key to obtaining meaningful data inthe open ocean as, often, the signals measured and samples col-lected are very close to the resolution and/or accuracy of theinstruments/sampling-technique. This is well illustrated by thedata of (Brewer et al., 1976), showing that the lowest and averagePM in the deep North Atlantic are on the order of 5–6 lg kg�1 and20 lg kg�1, respectively. Using large volume in situ pumps in theupper 500 m of the water column, Bishop and Edmond (1976)found particle concentrations to be 20–100% higher than previ-ously reported (Brewer et al., 1976), mostly due to adding thelarge-particle fraction. Gardner (1977) and Gardner et al. (1993)showed that inclusion of particles that have settled below theNiskin bottle sampling spigots increases particle concentrationsby 10–100%, which accounts well for the difference between bottleand pump concentrations reported by Bishop and Edmond (1976).

Assuming a conversion factor to beam attenuation of 0.8 m2 g�1

(based on the compilations of Gardner et al., 1985; Hill et al., 2011),the lowest and average cp at 660 nm are expected to be of the orderof 0.005 m�1 and 0.015 m�1, respectively.

The digital output of cp from 25 cm-long single band digitalbeam-transmissometers is reported to 0.001 m�1 resolution. Mon-itoring of analog output and digital count output from the sensorsduring laboratory bench top testing shows noise levels lower than0.0003 m�1. Acquisition of digital count data from the CTD permitsthis level of precision to be achieved. Aboard ship, along-track

Table 1Sensors and sensor systems recommendations for GEOTRACES cruises.

CTD-Rosette In-Line (shipboard flow through) Autonomous platforms

CommercialBeam-transmissometer Single band beam-transmissometer AUV and/or profiling floats with:Backscattering sensors Chlorophyll Fluorometer Beam-transmissometer, Backscattering or turbidity sensors.Turbidity sensors Spectral absorption and attenuation meter. Chlorophyll fluorometer.Chlorophyll fluorometers Flow-cytometerse

UVP a Imaging cytometersf

Automated particle imagersg

PrototypePIC sensorb Filtered/unfiltered optical propertiesc PIC sensor

PIC using acid-labile techniqued Optical (imaging)Sedimentation recordersb

a Picheral et al. (2010).b Bishop (2009).c Slade et al. (2010).d Balch and Utgoff (2009).e Dubelaar and Gerritzen (2000).f Olson and Sosik (2007).g Sieracki et al. (1998).

Table 2Summary table: Biogeochemical variables of particles and sensing technologies used to estimate them.

Variables Optical principle Sensor types

Mass concentration:Particle concentration (mass,

particulate organic carbon)Transmission, scattering, imaging Transmissometers, backscattering and side-scattering sensors,

camerasParticulate inorganic carbon Polarized transmission, acid- labile backscattering Custom transmissometers (PIC sensors), angular scattering sensors

with acid-injection.Phytoplankton pigments Fluorescence for chlorophyll-a like pigments and

Phycobiliprotein and particulate absorptionSingle or multi-band fluorometers and/or spectorphotometers

Particle size distributionNumber, area, volume distributions

as function of sizeImaging, near-forward scattering Submersible cameras, Near-forward scattering sensors, Imaging

cytometers, Automated particle imagers.Single size distribution parameter Slope of spectral particulate beam attenuation Multi spectral transmissometers

Fluctuation in optical properties Single wavelength transmissometers, backscattering andfluorometers

Particle compositionBulk tendency between organic and

inorganicRatio of backscattering to total scattering or beamattenuation

A backscattering sensor and spectral absorption and attenuationsensor or near-infra-red transmissometer

E. Boss et al. / Progress in Oceanography 133 (2015) 43–54 49

transmissometers fed with surface water can achieve 0.0005 m�1

accuracy by taking the difference between two successive measure-ments with the same instrument, one with and one without a0.2 lm filter at the intake of a flow sleeve, and averaging over aminute of independent measurements (Boss et al., 2007; Sladeet al., 2010).

Comparisons of sensors by Barnard and Claustre (personal com-munication 2012) and Bishop (CTD data R/V Point Sur, Feb 2012)demonstrated that (1) transmissometers are precise instruments(precision on the order of 0.001 m�1, and can be improved furtherby averaging), (2) transmissometers should be cleaned and cali-brated by the users close to the time of measurements to signifi-cantly increase their accuracy (GEOTRACES has published aprotocol on the appropriate procedures, see: http://www.geotrac-es.org/libraries/documents/Intercalibration/Cookbook.pdf), and (3)achieving highly accurate measurements in low particle concentra-tion waters, is still a major challenge. Details of these comparisonsare provided in Appendix B.

In terms of the back-scattering coefficient of particles, a mini-mum PM of 6 lg kg�1 with a mean around 20 lg kg�1 translatesto about 5 � 10�5 and 15 � 10�5 at 660 nm, where the uncertain-ties are on the order of 5 � 10�5m�1. To reach this level of accu-racy, however, dark current measurements have to bedetermined on a separate cast with black tape on the receiver(Twardowski et al., 2007). Twardowski et al. (2007) found that

between 300 and 500 m the difference between the backscatteringsignal at a red wavelength in the South Pacific ocean and the clean-est calibration waters was not significant.

Using turbidity meters it is possible to obtain higher resolutionthan with backscattering sensors. For example, Boss et al. (2008)obtained a minimal difference (7 counts) between the cleanestfield water and the dark current. The higher resolution with turbid-ity meters is most likely due to the larger (though less welldefined) sampling volume compared to backscattering sensors.The Seapoint turbidity meter, if measured to millivolt accuracy,has a resolution of 0.005 FTU (http://www.seapoint.com/stm.htm,equivalent to about 5 � 10�5 m�1 in backscattering units). Preci-sion can be improved using higher resolution analog to digital con-verters and by averaging.

Current technology to measure the absorption coefficient in-situ has accuracy on the order of 0.005 m�1 or 0.01 m�1 (for25 cm long WET-Labs’s ac-9 and ac-S respectively), when cali-brated with Nanopure clean water and for the red part of the spec-tra. Higher accuracy is possible if a calibration independenttechnique is used to obtain the particulate absorption(0.001 m�1, Boss et al., 2013), resulting in uncertainties in esti-mated chlorophyll on the order of 40% (Op. Cit.). Uncertaintiesassociated with the scattering correction increase in absolute valueat blue wavelengths and can be as high as 30% (Röttgers et al.,2013). When absorption by inorganic particles is important, signif-

50 E. Boss et al. / Progress in Oceanography 133 (2015) 43–54

icant errors in the particulate absorption coefficient in red wave-lengths are observed (Op. Cit.).

Because detection and analysis of particles vary among differentimaging systems, it is important to understand calibration proce-dures to ensure comparison of open ocean particle measurements.General calibration for optical instruments follows two criticalsteps: (1) calibration of the water volume for a single image and(2) calibration of the size of particle within the image. While par-ticle counting is very sensitive to step 1, particle sizing dependson step 2. The volume calibration can be performed in an aquariumfilled with seawater with the light source illuminating a calibrationsheet placed at different position in the field of light, and a digitalcamera recording the illuminated surface. Once all images arerecorded, the volume can be reconstructed in 3D. An example ofthe volume calibration process can be found in Picheral et al.(2010). The size calibration on the other hand can be done usingmainly 3 methods. For sufficiently large particles (>100 lm) cali-bration can be done inserting known ‘‘natural’’ targets into theimaging system prior to taking measurements (e.g. Picheral et al.2010). This allows calculating a calibration equation convertingpixel area to real surface area of particles. When looking at smallerparticles (<100 lm) inserting and tracking them in the light fieldbecomes impossible and calibrated beads or culture of phytoplank-ton cells are used to replace natural targets (Agrawal et al. 2008;Karp-Boss et al., 2007). Finally, cross-calibration of sensor outputwith a well characterized and calibrated sensor (‘gold standard’)can also be applied to insure inter-comparability of measurements.

Concentration has a critical impact on derived particle size dis-tributions. For example, imaging instruments have seldom beenused in oligotrophic systems where the principal technical chal-lenge resides in the low concentration of particles >500 lm. Sinceparticle concentration decreases approximately exponentially withsize, the illuminated volume must increase considerably in order todetect such particles and their statistics. For example, in the clear-est water of the Pacific south gyre, collecting 100 ml was found suf-ficient enough to detect particles up to 50 lm while 75 l ofseawater was necessary to detect particles as large as 1000 lm(Stemmann et al., 2008b).

From the discussion above it follows that the uncertainty associ-ated with obtaining optical properties with current technology maynot allow differentiation between a clear ocean water signal and ablank (the uncertainty in the signal is on the same order of magni-tude as its difference from the cleanest calibration water availableand heroic efforts need to be applied to do better). To improve onthis situation, instrumentation and protocols need to be developedthat can enable the robust determination of optical properties ofparticles in the deep ocean (e.g. using longer pathlength, strongersources, more sensitive/stable receivers, better signal stabilizationand calibration procedures). Specific calls for proposals by federalagencies (e.g. NSF-OTIC, NOPP or the SBIR programs in the USA) tar-geting the construction/refinement of novel/existing sensors thathave good signal/noise characteristics, and for their validation inthe clearest ocean waters, are needed to advance this field.

Conclusions and recommendations

Optical technologies can significantly aid the goals of the GEO-TRACES program in constraining material pools and processes asso-ciated with trace elements. While some of the optical sensors arelikely to be challenged in the clearest subsurface ocean waters, theywill provide very useful data near the ocean’s surface and bottomand will also be useful in detecting specific particle layers. GEO-TRACES should include particle optics during CTD/rosette castsand during in-situ filtration system deployments and treat such dataas core variables. The particulate samples, especially size fraction-

ated samples collected by in-situ filtration, when analyzed for majorconstituent phases, would provide an especially valuable resourcefor GEOTRACES and for interpretation of optical signals. Optical sen-sors must be capable of full water column deployment and thus notinterfere with cast operations. Further, at sea there must be scientificoversight of the optical data to ensure the quality of the data as it iscollected. It is further recommended that a CTD-based intercalibra-tion of multiple transmissometers, scattering and other candidateoptical sensors in low particle waters would benefit data qualityby providing the most stable instruments for major sections. TheGEOTRACES program would benefit from the addition of in-line(shipboard flow through) optical sensors to provide informationon horizontal distributions of particles within the mixed-layer.

Acknowledgements

This paper arose from the 3rd GEOTRACES Data-Model SynergyWorkshop in November 2011 at Universitat Autònoma de Barce-lona, Spain. Additional support for that workshop came from SCOR,through support to SCOR from the US NSF (Grant OCE-0938349and OCE-1243377) and from a US NSF grant to the US GEOTRACESProject Office (OCE-0850963) and US NSF support for J Bishop (OCE0964888). We thank Heidi Sosik and an anonymous reviewer forcomments and editing suggestions, which significantly improvedan earlier version of this manuscript.

A. Appendix: Repositories where optical data can be found

NSF funded, Biological and Chemical Data Management Office,bcodmo.org/.

ONR funded, Worldwide Ocean Optics Database (WOOD),wood.jhuapl.edu/wood/.

NASA’s SeaWIFS Bio-optical Archive and storage system (Sea-BASS), seabass.gsfc.nasa.gov/.

NOAA’s National Ocean Data Center, www.nodc.noaa.gov/.Texas A&M transmissometer and POC data sets from JGOFS,

WOCE, CLIVAR and others: ocean.tamu.edu/�pdgroup/SMP_prj/DataDir/SMP-data.html or

odv.awi.de/en/data/ocean/.

B. Appendix: Recent inter-comparison exercises involvingtransmissometers

In a comparison between ten new sensors deployed side-by-side (Barnard, Claustre, 2012 personal communication), it wasfound that if users did not clean and calibrate the sensors them-selves (that is relying on the manufacturer calibration and notcleaning the windows) differences between sensors readings couldbe as large as 0.05 m�1. Slight misalignment due to handling wheninstruments are shipped can cause significant deviation in read-ings. Conducting local calibration and cleaning the instrumentsimproved agreement between instruments to within 0.01 m�1.Forcing all instruments to agree on the value at depth resulted ina difference between instruments that was on the order of0.001 m�1.

Four transmissometers of two different designs (WET-Labs’ C-STAR analog and digital instruments (CST 391 and CST 1450) andtwo prototype neutrally buoyant instruments, WET-Labs’ POC017,POC018) were simultaneously profiled in deep California Currentwaters over 5 days at the same location (Bishop, unpublished data,2012). Published GEOTRACES protocols for transmissometers werefollowed. Pooled sensor results for a low variability water layerbetween 1400 and 1600 m gave an average beam attenuation coef-ficient value of 0.0085 (s.d. 0.0033) m�1 (Fig. A1, Lerner et al., 2013).In contrast, each individual instrument repeated the 1400–1600 m

Fig. A1. Result of a test of 4 transmissometers R/V Point Sur, Feb 2012. Datarepresent the averages of beam attenuation coefficient over 3 days in a lowvariability layer from 1400–1600 m near Monterey Bay, CA. UPPER vs. LOWERpanels are data from CST391 with and without a correction applied to account forthermal effects on transmissometer beam intensity. This correction adjustmentresulted in an improved agreement between transmissometers from ±0.0033 m�1

to ±0.0021 m�1.

E. Boss et al. / Progress in Oceanography 133 (2015) 43–54 51

beam attenuation coefficient value to a precision ofs.d = 0.0009 m�1. The largest contributor to the absolute disagree-ment of the 4 sensors was that each had a different and finite resid-ual signal response to temperature, which although falling withinmanufacturer’s specifications, is repeatable and can be compen-sated for during data processing. We were able to correct data fromthe older – less accurate analog sensor CST391 for temperatureeffects and brought all results into agreement of ±0.0022 m�1. Man-ufacturer calibration of the sensor in particle free water has beenshown to be reproducible. The residual thermal sensitivity issuesof transmissometers can be solved.

References

Aas, E., 1996. Refractive index of phytoplankton derived from its metabolitecomposition. Journal of Plankton Research 18, 2223–2249.

Agrawal, Y.C., Pottsmith, H.C., 2000. Instruments for particle size and settlingvelocity observations in sediment transport. Marine Geology 168, 89–114.

Agrawal, Y.C., Whitmire, A., Mikkelsen, O.A., Pottsmith, H.C., 2008. Light scatteringby random shaped particles and consequences on measuring suspendedsediments by laser diffraction. Journal of Geophysical Research 113, C04023.http://dx.doi.org/10.1029/2007JC004403.

Alexander, S., Anderson, S., Hendry, D.C., Hobson, P.R., Lampitt, R.S., Lucas-Leclin, B.,Nareid, H., Nebrensky, J.J., Player, M.A., Saw, K., Tipping, K., Watson, J., 2000.HoloCam: a subsea holographic camera for recording marine organisms andparticles. Optical Diagnostics for Industrial Applications 4076, 111–119.

Alldredge, A., 1998. The carbon, nitrogen and mass content of marine snow as afunction of aggregate size. Deep-Sea Research Part I-Oceanographic ResearchPapers 45, 529–541.

Alldredge, A.L., Jackson, G.A., 1995. Aggregation in marine systems. Deep-SeaResearch Part-II 42, 1–7.

Alldredge, A.L., Passow, U., Logan, B.E., 1993. The abundance and significance of aclass of large, transparent organic particles in the ocean. Deep-Sea Research,Part-I 40, 1131–1140.

Alvain, S., Moulin, C., Dandonneau, Y., Breon, F.M., 2005. Remote sensing ofphytoplankton groups in case 1 waters from global SeaWiFS imagery. Deep-SeaResearch Part I-Oceanographic Research Papers 52, 1989–2004.

Anderson, R.F., Hayes, C.T., 2015. Characterizing marine particles and their impacton biogeochemical cycles in the GEOTRACES program. Progress InOceanography 133, 1–5.

Asper, V.L., 1987. Measuring the flux and sinking speed of marine snow aggregates.Deep-Sea Research Part a-Oceanographic Research Papers 34, 1–17.

Asper, V.L., Honjo, S., Orsi, T.H., 1992. Distribution and transport of marine snowaggregates in the Panama Basin. Deep-Sea Research 39 (6A), 939–952.

Babin, M., Morel, A., Fournier-Sicre, V., Fell, F., Stramski, D., 2003. Light scatteringproperties of marine particles in coastal and open ocean waters as related to theparticle mass concentration. Limnology and Oceanography48, 843–859.

Baker, E.T., Tennant, D.A., Feely, R.A., Lebon, G.T., Walker, G.T., 2001. Field andlaboratory studies on the effect of particle size and composition on opticalbackscattering measurements in hydrothermal plumes. Deep-Sea Research,Part-I 48, 593–604.

Balch, W.M., Utgoff, P.E., 2009. Potential interactions among ocean acidification,coccolithophores, and the optical properties of seawater. Oceanography 22,146–159.

Balch, W.M., Drapeau, D.T., Fritz, J.J., Bowler, B.C., Nolan, J., 2001. Opticalbackscattering in the Arabian Sea – continuous underway measurements ofparticulate inorganic and organic carbon. Deep-Sea Research Part I-Oceanographic Research Papers 48, 2423–2452.

Balch, W., Drapeau, D., Bowler, B., Booth, E., 2007. Prediction of pelagic calcificationrates using satellite measurements. Deep-Sea Research Part Ii-Topical Studies inOceanography 54, 478–495.

Barnard, A.H., Pegau, W.S., Zaneveld, J.R.V., 1998. Global relationships of theinherent optical properties of the oceans. Journal of Geophysical Research-Oceans 103, 24955–24968.

Behrenfeld, M.J., Boss, E., 2006. Beam attenuation and chlorophyll concentration asalternative optical indices of phytoplankton biomass. Journal of MarineResearch 64, 431–451.

Benfield, M.C., Grosjean, P., Culverhouse, P.F., Irigoien, X., Sieracki, M.E., Lopez-Urrutia, A., Dam, H.G., Hu, Q., Davis, C.S., Hansen, A., Pilskaln, C.H., Riseman,E.M., Schultz, H., Utgoff, P.E., Gorsky, G., 2007. RAPID research on automatedplankton identification. Oceanography 20, 172–187.

Bishop, J.K.B., 1986. The correction and suspended particulate matter calibration ofsea tech transmissometer data. Deep-Sea Research Part a-OceanographicResearch Papers 33, 121–134.

Bishop, J.K.B., 1999. Transmissometer measurement of POC. Deep-Sea Research PartI-Oceanographic Research Papers 46, 353–369.

Bishop, J.K.B., 2009. Autonomous observations of the ocean biological carbon pump.Oceanography 22, 182–193.

Bishop, J.K.B., Edmond, J.M., 1976. New large volume filtration system for samplingof oceanic particulate matter. Journal of Marine Research 34, 181–198.

Bishop, J.K.B., Joyce, T.M., 1986. Spatial distributions and variability of suspendedparticulate matter in warm-core ring 82b. Deep-Sea Research Part a-Oceanographic Research Papers 33, 1741–1760.

Bishop, J.K.B., Wood, T.J., 2008. Particulate matter chemistry and dynamics in thetwilight zone at VERTIGO ALOHA and K2 sites. Deep Sea Research Part I:Oceanographic Research Papers 55, 1684–1706.

Bishop, J.K.B., Wood, T.J., 2009. Year round observations of carbon biomass and fluxvariability in the southern ocean. Global Biogeochemical Cycles 23, GB2019.http://dx.doi.org/10.1029/2008GB003206.

Bishop, J.K.B., Smith, R.C., Baker, K.S., 1992. Springtime distributions and variabilityof biogenic particulate matter in Gulf-Stream warm-core ring 82b andsurrounding Nw Atlantic Waters. Deep-Sea Research Part a-OceanographicResearch Papers 39, S295–S325.

Bishop, J.K.B., Calvert, S.E., Soon, M.Y.S., 1999. Spatial and temporal variability ofPOC in the northeast Subarctic Pacific. Deep-Sea Research Part Ii-Topical Studiesin Oceanography 46, 2699–2733.

Bishop, J.K.B., Davis, R.E., Sherman, J.T., 2002. Robotic observations of dust stormenhancement of carbon biomass in the North Pacific. Science 298, 817–821.

Bishop, J.K.B., Wood, T.J., Davis, R.E., Sherman, J.T., 2004. Robotic observations ofenhanced carbon biomass and export at 55 degrees S during SOFeX. Science304, 417–420.

Bishop, J.K.B., Lam, P.J., Wood, T.J., 2012. Getting good particles: accurate samplingof particles by large volume in-situ filtration. Limnology and OceanographyMethods 10, 681–710.

Boss, E., Pegau, W.S., Gardner, W.D., Zaneveld, J.R.V., Barnard, A.H., Twardowski,M.S., Chang, G.C., Dickey, T.D., 2001a. Spectral particulate attenuation andparticle size distribution in the bottom boundary layer of a continental shelf.Journal of Geophysical Research-Oceans 106, 9509–9516.

Boss, E., Twardowski, M.S., Herring, S., 2001b. Shape of the particulate beamattenuation spectrum and its inversion to obtain the shape of the particulatesize distribution. Applied Optics 40, 4885–4893.

Boss, E.S., Collier, R., Larson, G., Fennel, K., Pegau, W.S., 2007. Measurements ofspectral optical properties and their relation to biogeochemical variables andprocesses in Crater Lake, Crater Lake National Park, OR. Hydrobiologia 574,149–159.

Boss, E., Swift, D., Taylor, L., Brickley, P., Zaneveld, R., Riser, S., Perry, M.J., Strutton,P.G., 2008. Observations of pigment and particle distributions in the westernNorth Atlantic from an autonomous float and ocean color satellite. Limnologyand Oceanography 53, 2112–2122.

Boss, E., Slade, W., Hill, P., 2009a. Effect of particulate aggregation in aquaticenvironments on the beam attenuation and its utility as a proxy for particulatemass. Optics Express 17, 9408–9420.

52 E. Boss et al. / Progress in Oceanography 133 (2015) 43–54

Boss, E., Slade, W.H., Behrenfeld, M., Dall’Olmo, G., 2009b. Acceptance angle effectson the beam attenuation in the ocean. Optics Express 17, 1535–1550.

Boss, E., Taylor, L., Gilbert, S., Gundersen, K., Hawley, N., Janzen, C., Johengen, T.,Purcell, H., Robertson, C., Schar, D.W.H., Smith, G.J., Tamburri, M.N., 2009c.Comparison of inherent optical properties as a surrogate for particulate matterconcentration in coastal waters. Limnology and Oceanography-Methods 7, 803–810.

Boss, E., Picheral, M., Leeuw, T., Chase, A., Karsenti, E., Gorsky, G., Taylor, L., Slade,W., Ras, J., Claustre, H., 2013. The characteristics of particulate absorption,scattering and attenuation coefficients in the surface ocean. Contribution of theTara Oceans expedition, Methods in Oceanography 7, 52–62.

Braithwaite, K.M., Bowers, D.G., Smith, W.A.M.N., Graham, G.W., 2012. Controls onfloc growth in an energetic tidal channel. Journal of Geophysical Research-Oceans 117. http://dx.doi.org/10.1029/2011JC007094.

Brewer, P.G., Spencer, D.W., Biscaye, P.E., Hanley, A., Sachs, P.L., Smith, C.L., Kadar, S.,Fredericks, J., 1976. The distribution of particulate matter in the Atlantic Ocean.Earth and Planetary Science Letters 32, 393–402.

Brewin, R.J.W., Hardman-Mountford, N.J., Lavender, S.J., Raitsos, D.E., Hirata, T., Uitz,J., Devred, E., Bricaud, A., Ciotti, A., Gentili, B., 2011. An intercomparison of bio-optical techniques for detecting phytoplankton size class from satellite remotesensing. Remote Sensing of Environment 115, 325–339. http://dx.doi.org/10.1016/j.rse.2010.09.004.

Briggs, N.T., Slade, W.H., Boss, E., Perry, M.J., 2013. Method for estimating meanparticle size from high-frequency fluctuations in beam attenuation or scatteringmeasurements. Applied Optics 52, 6710–6725.

Burd, A.B., Jackson, G.A., 2009. Particle aggregation. Annual Review in MarineSciences 1, 65–90.

Burd, A.B., Jackson, G.A., Moran, S.B., 2007. The role of the particle size spectrum inestimating POC fluxes from disequilibrium. Deep Sea Research Part I:Oceanographic Research Papers 54, 897–918.

Cetinic, I., Perry, M.J., Briggs, N.T., Kallin, E., D’Asaro, E.A., Lee, C.M., 2012. Particulateorganic carbon and inherent optical properties during 2008 North AtlanticBloom Experiment. Journal of Geophysical Research-Oceans 117, C06028.http://dx.doi.org/10.1029/2011JC007771.

Chase, A., Boss, E., Zaneveld, R., Bricaud, A., Claustre, H., Ras, J., Dall’Olmo, G.,Westberry, T., 2013. Decomposition of in situ particulate absorption spectra.Methods in Oceanography 7, 110–124. http://dx.doi.org/10.1016/j.mio.2014.02.002.

Checkley, D.M., Davis, R.E., Herman, A.W., Jackson, G.A., Beanlands, B., Regier, L.A.,2008. Assessing plankton and other particles in situ with the SOLOPC.Limnology and Oceanography 53, 2123–2136.

Cowen, R.K., Guigand, C.M., 2008. In situ ichthyoplankton imaging system (ISIIS):system design and preliminary results. Limnology and Oceanography-Methods6, 126–132.

Davis, R.F., Moore, C.C., Zaneveld, J.R.V., Napp, J.M., 1997. Reducing the effects offouling on chlorophyll estimates derived from long-term deployments ofoptical instruments. Journal of Geophysical Research 102, 5851–5855.

Davis, C.S., Thwaites, F.T., Gallager, S.M., Hu, Q., 2005. A three-axis fast-tow digitalVideo Plankton Recorder for rapid surveys of plankton taxa and hydrography.Limnology and Oceanography-Methods 3, 59–74.

Deng, F., Thomas, A.L., Rijkenberg, M.J.A., Henderson, G.M., 2014. Controls onseawater 231Pa, 230Th and 232Th concentrations along the flow paths of deepwaters in the Southwest Atlantic. Earth and Planetary Science Letters 390, 93–102.

Dickey, T., Lewis, M., Chang, G., 2006. Optical oceanography: recent advances andfuture directions using global remote sensing and in situ observations. Reviewsof Geophysics 44, RG 1001. http://dx.doi.org/10.1029/2003RG000148.

Dickey, T.D., Itsweire, E.C., Moline, M., Perry, M.J., 2008. Introduction to theLimnology and Oceanography Special Issue on Autonomous and LagrangianPlatforms and Sensors (ALPS). Limnology and Oceanography 53, 2057–2061.

Diercks, A.R., Asper, V.L., 1997. In situ settling speeds of marine snow aggregatesbelow the mixed layer: Black Sea and Gulf of Mexico. Deep-Sea Research II 44,385–397.

Dubelaar, G.B.J., Gerritzen, P.L., 2000. CytoBuoy: a step forward towards using flowcytometry in operational oceanography. Sciancias Marinas 64, 255–265.

Duysens, L.M.N., 1956. The flattening of the absorption spectrum of suspensions, ascompared to that of solutions. Biochimica et Biophysica Acta 19, 1–12.

Estapa, M.L., Boss, E., Mayer, L.M., Roesler, C.S., 2012. Role of iron and organic carbonin mass-specific light absorption by particulate matter from Louisiana coastalwaters. Limnology and Oceanography 57, 97–112.

Fournier, G. R., Sanjuan-Calzado, V., Trees, C., 2014. Implications of a new phasefunction for autonomous underwater imaging. In: Proc. SPIE 9111, OceanSensing and Monitoring VI, 911119, http://dx.doi.org/10.1117/12.2053878.

Fry, E.S., Kattawar, G.W., Pope, R.M., 1992. Integrating cavity absorption meter.Applied Optics 31, 2055–2065.

Gardner, W.D., 1977. Incomplete extraction of rapidly settling particles from watersamplers. Limnology and Oceanography 22, 764–768.

Gardner, W.D., 1989a. Baltimore canyon as a modern conduit of sediment to thedeep-sea. Deep-Sea Research Part a-Oceanographic Research Papers 36, 323–358.

Gardner, W.D., 1989b. Periodic resuspension in Baltimore Canyon by focusing ofinternal waves. Journal of Geophysical Research 94, 18185–18194.

Gardner, W.D., Walsh, I.D., 1990. Distribution of macroaggregates and fine-grainedparticles across a continental-margin and their potential role in fluxes. Deep-Sea Research Part a-Oceanographic Research Papers 37, 401–411.

Gardner, W.D., Biscaye, P.E., Zaneveld, J.R.V., Richardson, M.J., 1985. Calibration andcomparison of the LDGO nephelometer and the OSU transmissometer on theNova Scotian rise. Marine Geology 66, 323–344.

Gardner, W.D., Walsh, I.D., Richardson, M.J., 1993. Biophysical forcing of particleproduction and distribution during a spring bloom in the North Atlantic. DeepSea Research Part II: Topical Studies in Oceanography 40, 171–195.

Gardner, W.D., Richardson, M.J., Smith, W.O., 2000. Seasonal patterns of watercolumn particulate organic carbon and fluxes in the Ross Sea, Antarctica. Deep-Sea Research II 47, 3423–3449.

Gardner, W.D., Blakey, J.C., Walsh, I.D., Richardson, M.J., Pegau, S., Zaneveld, J.R.V.,Roesler, C., Gregg, M.C., MacKinnon, J.A., Sosik, H.M., Williams III, A.J., 2001.Optics, particles, stratification and storms on the New England continentalshelf. Journal of Geophysical Research 106, 9473–9497.

Gardner, W.D., Richardson, M.J., Carlson, C.A., Hansell, D., Mishonov, A.V., 2003.Determining true particulate organic carbon: bottles, pumps andmethodologies. Deep-Sea Research Part Ii-Topical Studies in Oceanography50, 655–674.

Gardner, W.D., Richardson, M.J., Mishonov, A.V., Biscaye, P.E., 2014. GlobalDistribution And Intensity Of Deep-Water Benthic Nepheloid Layers – WhatSatellites, Floats And Gliders Don’t See. Ocean Sciences Meeting, Hawaii.

Goldthwait, S.A., Alldredge, A.L., 2006. An investigation of diel synchronicitybetween water column marine snow concentration and the flux of organicmatter in the Santa Barbara Channel, California. Deep-Sea Research Part I-Oceanographic Research Papers 53, 485–505.

Gonzalez-Quiros, R., Checkley, D.M., 2006. Occurrence of fragile particles inferredfrom optical plankton counters used in situ and to analyze net samples collectedsimultaneously. Journal of Geophysical Research – Oceans 111, 5–6.

Gorsky, G., Aldorf, C., Kage, M., Picheral, M., Garcia, Y., Favole, J., 1992. Vertical-distribution of suspended aggregates determined by a new underwater videoprofiler. Annales De L Institut Oceanographique 68, 275–280.

Guay, C.K.H., Bishop, J.K.B., 2002. A rapid birefringence method for measuringsuspended CaCO3 concentrations in seawater. Deep-Sea Research Part I-Oceanographic Research Papers 49, 197–210.

Guidi, L., Jackson, G.A., Stemmann, L., Miquel, J.C., Picheral, M., Gorsky, G., 2008.Relationship between particle size distribution and flux in the mesopelagiczone. Deep-Sea Research Part I-Oceanographic Research Papers 55, 1364–1374.

Guidi, L., Stemmann, L., Jackson, G.A., Ibanez, F., Claustre, H., Legendre, L., Picheral,M., Gorsky, G., 2009. Effects of phytoplankton community on production, sizeand export of large aggregates: a world-ocean analysis. Limnology andOceanography 54, 1951–1963.

Guidi, L., Calil, P.H.R., Duhamel, S., Björkman, K.M., Doney, S.C., Jackson, G.A., Li, B.,Church, M.J., Tozzi, S., Kolber, Z.S., Richards, K.J., Fong, A.A., Letelier, R.M.,Gorsky, G., Stemmann, L., Karl, D.M., 2012. Does eddy-eddy interaction controlsurface phytoplankton distribution and carbon export in the North PacificSubtropical Gyre? Journal of Geophysical Research 117, G02024.

Hayes, C.T., Anderson, R.F., Fleisher, M.Q., Huang, K.F., Robinson, L.F., Lu, Y., Cheng,H., Edwards, R.L., Moran, S.B., in press. 230Th and 231Pa on GEOTRACES GA03, theU.S. GEOTRACES North Atlantic transect and implications for modern andpaleoceanographic chemical fluxes. Deep Sea Research Part II: Topical Studies inOceanography, http://dx.doi.org/10.1016/j.dsr2.2014.07.007i.

Hill, P.S., Boss, E., Newgard, J.P., Law, B.A., Milligan, T.G., 2011. Observations of thesensitivity of beam attenuation to particle size in a coastal bottom boundarylayer. Journal of Geophysical Research-Oceans 116, C02023. http://dx.doi.org/10.1029/2010JC006539.

Hobson, P.R., Krantz, E.P., Lampitt, R.S., Rogerson, A., Watson, J., 1997. A preliminarystudy of the distribution of plankton using hologrammetry. Optics and LaserTechnology 29, 25–33.

Holser, R.R., Goni, M.A., Hales, B., 2011. Design and application of a semi-automatedfiltration system to study the distribution of particulate organic carbon in thewater column of a coastal upwelling system. Marine Chemistry 123 (1–4), 67–77. http://dx.doi.org/10.1016/j.marchem.2010.10.001.

Honjo, S., Doherty, K.W., Agrawal, Y.C., Asper, V.L., 1984. Direct optical assessmentof large amorphous aggregates (Marine Snow) in the deep ocean. Deep-SeaResearch Part a-Oceanographic Research Papers 31, 67–76.

IOCCG, 2006. Remote sensing of inherent optical properties: fundamentals, tests ofalgorithms, and applications. In: Lee, Z.P. (Ed.), Reports of the InternationalOcean-Colour Coordinating Group, No. 5, IOCCG, Dartmouth, Canada.

IOCCG, 2011. Bio-Optical Sensors on Argo Floats. Claustre, H. (ed.), Reports of theInternational Ocean-Colour Coordinating Group, No. 11, IOCCG, Dartmouth,Canada.

IOCCG, 2014. Phytoplankton Functional Types from Space. In: Sathyendranath, S.(Ed.), Reports of the International Ocean-Colour Coordinating Group, No. 15,IOCCG, Dartmouth, Canada.

Iversen, M.H., Nowald, N., Ploug, H., Jackson, G.A., Fischer, G., 2010. High resolutionprofiles of vertical particulate organic matter export off Cape Blanc, Mauritania:degradation processes and ballasting effects. Deep-Sea Research Part I-Oceanographic Research Papers 57, 771–784.

Jackson, G., Burd, A., 2002. A model for the distribution of particle flux in the mid-water column controlled by subsurface biotic interactions. Deep-Sea Research II49, 193–217.

Jackson, G.A., Checkley, D.M., 2011. Particle size distributions in the upper 100 mwater column and their implications for animal feeding in the plankton. Deep-Sea Research Part I: Oceanographic Research Papers 58, 283–297.

Jackson, G.A., Maffione, R., Costello, D.K., Alldredge, A.L., Logan, B.E., Dam, H.G.,1997. Particle size spectra between 1 lm and 1 cm at Monterey Bay determined

E. Boss et al. / Progress in Oceanography 133 (2015) 43–54 53

using multiple instruments. Deep Sea Research Part I: Oceanographic ResearchPapers 44, 1739–1767.

Jeandel, C., Rutgers van der Loeff, M., Kretschmer, S., Lam, P.J., Sherrell, R.M., Roy-Barman, M., German, C.R., Dehairs, F., 2015. What did we learn on the oceanicparticle dynamic from the GEOSECS-JGOFS times? Progress in Oceanography133, 6–16.

Jonasz, M., Fournier, G., 2007. Light Scattering by Particles in Water: Theoretical andExperimental Foundations. Elsevier, Amsterdam.

Jouandet, M.P., Trull, T.W., Guidi, L., Picheral, M., Ebersbach, F., Stemmann, L., Blain,S., 2011. Optical imaging of mesopelagic particles indicates deep carbon fluxbeneath a natural iron-fertilized bloom in the Southern Ocean. Limnology andOceanography 56, 1130–1140.

Karageorgis, A., Gardner, W.D., Georgopoulos, D., Mishonov, A.V., Krasakopoulou, E.,Anagnostou, C., 2008. Particle dynamics in the Eastern Mediterranean Sea: asynthesis based on light transmission, PMC, and POC archives (1991–2001).Deep-Sea Research I 55, 177–202. http://dx.doi.org/10.1016/j.dsr.2007.11.002.

Karageorgis, A.P., Gardner, W.D., Mikkelsen, O.A., Georgopoulos, D., Ogston, A.S.,Assimakopoulou, G., Krasakopoulou, E., Oaie, Gh., Secrieru, D., Kanellopoulos,Th.D., Pagou, K., Anagnostou, Ch., Papathanassiou, E., 2014. Particle sources overthe Danube River Delta, Black Sea based on distribution, composition andsize using optics, imaging and bulk analyses. Journal of Marine Systems 131,74–90.

Karp-Boss, L., Wheeler, P.A., Hales, B., Covert, P., 2004. Distributions and variabilityof particulate organic matter in a coastal upwelling system. Journal ofGeophysical Research 109, C09010. http://dx.doi.org/10.1029/2003JC002184.

Karp-Boss, L., Azevedo, L., Boss, E., 2007. LISST-100 measurements of phytoplanktonsize distribution: evaluation of the effects of cell shape. Limnology andOceanography-Methods 5, 396–406.

Katz, J., Donaghay, P., Zhang, J., King, S., Russell, K., 1999. Submersible holocamerafor detection of particle characteristics and motions in the ocean. Deep-SeaResearch Part I 46, 1455–1481.

Kostadinov, T.S., Siegel, D.A., Maritorena, S., 2009. Retrieval of the particle sizedistribution from satellite ocean color observations. Journal of GeophysicalResearch-Oceans 114, C09015. http://dx.doi.org/10.1029/2009JC005303.

Kostadinov, T.S., Siegel, D.A., Maritorena, S., 2010. Global variability ofphytoplankton functional types from space: assessment via the particle sizedistribution. Biogeosciences 7, 3239–3257.

Lam, P.J., Twining, B.S., Jeandel, C., Roychoudhury, A.N., Resing, J., Geibert, W.,Santschi, P., Anderson, R.F., 2015. Methods for analyzing the concentration andspeciation of major and trace elements in marine particles. Progress inOceanography 133, 32–42.

Lerner, P., Bishop, J.K.B., Strubhar W.D., Bernard, A., Moore, C., 2013.Transmissometer measurement of particle beam attenuation coefficient. ALSOOcean Sciences Meeting Session SS53-380. New Orleans Feb 16–21 2013.

Leymarie, E., Doxaran, D., Babin, M., 2010. Uncertainties associated tomeasurements of inherent optical properties in natural waters. Applied Optics49, 5415–5436.

Loisel, H., Nicolas, J.-M., Sciandra, A., Stramski, D., Poteau, A., 2006. Spectraldependency of optical backscattering by marine particles from satellite remotesensing of the global ocean. Journal of Geophysical Research 111, C09024.http://dx.doi.org/10.1029/2005JC003367.

Loisel, H., Meriaux, X., Berthon, J.F., Poteau, A., 2007. Investigation of the opticalbackscattering to scattering ratio of marine particles in relation to theirbiogeochemical composition in the eastern English Channel and southern NorthSea. Limnology and Oceanography 52, 739–752.

MacIntyre, H. L., Lawrenz, E., Richardson, T.L., 2010. Taxonomic Discrimination ofPhytoplankton by Spectral Fluorescence. Ch. 7 In: D.J. Chlorophyll aFluorescence in Aquatic Sciences: Methods and Applications, Developments inApplied Phycology 4, Suggett et al. (Eds.), Springer http://dx.doi.org/10.1007/978-90-481-9268-7_7.

McDonnell, A.M.P., Buesseler, K.O., 2010. Variability in the average sinkingvelocities of marine particles. Limnology and Oceanography 55, 2085–2096.

McDonnell, A.M.P., Lamborg, C.H., Buesseler, K.O., Lam, P.J., Sanders, R., Smith, H.,Sargent, E.C., Riley, J.S., Lampitt, R.S., Marsay, C. and Bishop, J.K.B., 2015. Asummary of techniques for the collection of marine particles. Progress inOceanography 133, 17–31.

McKee, D., Chami, M., Brown, I., Sanjuan Calzado, V., Doxaran, D., Cunningham, A.,2009. Role of measurement uncertainties in observed variability in the spectralbackscattering ratio: a case study in mineral-rich coastal waters. Applied Optics48, 4663–4675.

Moisan, J.R., Moisan, T.A.H., Linkswiler, M.A., 2011. An inverse modeling approach toestimating phytoplankton pigment concentrations from phytoplanktonabsorption spectra. Journal of Geophysical Research-Oceans 116, C09018.http://dx.doi.org/10.1029/2010JC006786.

Moran, S.B., Charette, M.A., Pike, S.M., Wicklund, C.A., 1999. Differences in seawaterparticulate organic carbon concentration in samples collected using small- andlarge volume methods: the importance of DOC adsorption to the filter blank.Marine Chemistry 67, 33–42.

Neukermans, G., Loisel, H., Meriaux, X., Astoreca, R., Mckee, D., 2012. In situvariability of mass-specific beam attenuation and backscattering of marineparticles with respect to particle size, density, and composition. Limnology andOceanography 57, 124–144.

Olson, R.J., Sosik, H.M., 2007. A submersible imaging-in-flow instrument to analyzenano-and microplankton: imaging FlowCytobot. Limnology and Oceanography-Methods 5, 195–203.

O’Reilly, J.E., Maritorena, S., Mitchell, B.G., Siegel, D.A., Carder, K.L., Garver, S.A.,Kahru, M., McClain, C., 1998. Ocean color chlorophyll algorithms for SeaWiFS.Journal of Geophysical Research-Oceans 103, 24937–24953.

Peterson, M.L., Wakeham, S.G., Askea, M.A., Miquel, J.C., 2005. Novel techniques forcollection of sinking particles in the ocean and determining their settling rates.Limnology and Oceanography: Methods 3, 520–532.

Pfitsch, D.W., Malkiel, E., Takagi, M., Ronzhes, Y., King, S., Sheng, J., Katz, J., 2007.Analysis of in-situ microscopic organism behavior in data acquired using a free-drifting submersible holographic imaging system. Oceans 1–5, 583–590.

Picheral, M., Guidi, L., Stemmann, L., Karl, D.M., Iddaoud, G., Gorsky, G., 2010. TheUnderwater Vision Profiler 5: an advanced instrument for high spatialresolution studies of particle size spectra and zooplankton. Limnology andOceanography-Methods 8, 462–473.

Reynolds, R.A., Stramski, D., Wright, V.M., Wozniak, S.B., 2010. Measurements andcharacterization of particle size distributions in coastal waters. Journal ofGeophysical Research-Oceans 115. http://dx.doi.org/10.1029/2009JC005930.

Richardson, M. J., Gardner, W.D., Berglund, B.L., and Walsh, I.D., 1990. Globaldistribution of biogenic and particulate matter: What satellites can’t see, buttransmissometers can. EOS, Trans. Amer. Geophys. Union, 71, 136.

Richardson, M.J., Weatherly, G.L., Gardner, W.D., 1993. Benthic storms in theArgentine Basin. Deep-Sea Research Part Ii-Topical Studies in Oceanography 40,975–987.

Röttgers, R., McKee, D., Wozniak, S.B., 2013. Evaluation of scatter corrections for ac-9 absorption measurements in coastal waters. Methods in Oceangraphy 7, 21–39.

Roullier, F., Berline, L., Guidi, L., Sciandra, A., Durrieu de Madron, X., Picheral, M.,Pesant, S., Stemmann, L., 2014. Particles size distribution and carbon flux acrossthe Arabian Sea Ocean Oxygen Minimum Zone. Biogeosciences 11, 4541–4557.

Samson, S., Hopkins, T., Remsen, A., Langebrake, L., Sutton, T., Patten, J., 2001. Asystem for high-resolution zooplankton imaging. Ieee Journal of OceanicEngineering 26, 671–676.

Shifrin, K.S., 1988. The method of fluctuations. In: Physical Optics of Ocean Water.American Institute of Physics, 1988, pp. 224–232.

Sieracki, C.K., Sieracki, M.E., Yentsch, C.S., 1998. An imaging-in-flow system forautomated analysis of marine microplankton. Marine Ecolology Progress Series168, 285–296.

Slade, W.H., Boss, E., Dall’Olmo, G., Langner, M.R., Loftin, J., Behrenfeld, M.J., Roesler,C., Westberry, T.K., 2010. Underway and moored methods for improvingaccuracy in measurement of spectral particulate absorption and attenuation.Journal of Atmospheric and Oceanic Technology 27 (10), 1733–1746.

Slade, W.H., Boss, E., Russo, C., 2011. Effects of particle aggregation anddisaggregation on their inherent optical properties. Optics Express 19, 7945–7959.

Son, Y.B., Gardner, W.D., Mishonov, A.V., Richardson, M.J., 2009. Multispectralremote-sensing algorithms for particulate organic carbon (POC): the Gulf ofMexico. Remote Sensing of Environment 113, 50–61.

Sosik, H.M., 2008. Characterizing seawater constituents from optical properties. In:Babin, M., Roesler, C.S., Cullen, J.J. (Eds.), Real-time Coastal Observing Systemsfor Ecosystem Dynamics and Harmful Algal Blooms: Theory, instrumentationand modelling. UNESCO, p. 281–329.

Stemmann, L., Boss, E., 2012. Plankton and particle size and packaging: fromdetermining optical properties to driving the biological pump. Annual Review ofMarine Science 4 (4), 263–290.

Stemmann, L., Jackson, G.A., Gorsky, G., 2004a. A vertical model of particle sizedistributions and fluxes in the midwater column that includes biological andphysical processes – Part II: application to a three year survey in the NWMediterranean Sea. Deep-Sea Research Part I-Oceanographic Research Papers51, 885–908.

Stemmann, L., Jackson, G.A., Ianson, D., 2004b. A vertical model of particle sizedistributions and fluxes in the midwater column that includes biological andphysical processes – Part I: model formulation. Deep-Sea Research Part I-Oceanographic Research Papers 51, 865–884.

Stemmann, L., Youngbluth, M., Robert, K., Hosia, A., Picheral, M., Paterson, H.,Ibanez, F., Guidi, L., Lombard, F., Gorsky, G., 2008a. Global zoogeography offragile macrozooplankton in the upper 100–1000 m inferred from theunderwater video profiler. Ices Journal of Marine Science 65, 433–442.

Stemmann, L., Eloire, D., Sciandra, A., Jackson, G.A., Guidi, L., Picheral, M., Gorsky, G.,2008b. Volume distribution for particles between 3.5 to 2000 lm in the upper200 m region of the South Pacific Gyre. Biogeosciences 5, 299–310.

Stramski, D., Reynolds, R.A., Kahru, M., Mitchell, B.G., 1999. Estimation of particulateorganic carbon in the ocean from satellite remote sensing. Science 285, 239–242.

Stramski, D., Reynolds, R.A., Babin, M., Kaczmarek, S., Lewis, M.R., Rottgers, R.,Sciandra, A., Stramska, M., Twardowski, M.S., Franz, B.A., Claustre, H., 2008.Relationships between the surface concentration of particulate organic carbonand optical properties in the eastern South Pacific and eastern Atlantic Oceans.Biogeosciences 5, 171–201.

Sullivan, J., Twardowski, M., Zaneveld, J.R.V., Moore, C., 2013. Measuring opticalbackscattering in water. In: Kokhanovsky, A. (Ed.), Light Scattering Reviews 7:Radiative Transfer and Optical Properties of Atmosphere and UnderlyingSurface, Springer Praxis Books, pp. 189–224 http://dx.doi.org/10.1007/978-3-642-21907-8_6.

Twardowski, M.S., Boss, E., Macdonald, J.B., Pegau, W.S., Barnard, A.H., Zaneveld,J.R.V., 2001. A model for estimating bulk refractive index from the opticalbackscattering ratio and the implications for understanding particle

54 E. Boss et al. / Progress in Oceanography 133 (2015) 43–54

composition in case I and case II waters. Journal of Geophysical Research-Oceans 106, 14129–14142.

Twardowski, M.S., Claustre, H., Freeman, S.A., Stramski, D., Huot, Y., 2007. Opticalbackscattering properties of the ‘clearest’ natural waters. Biogeosciences 4,1041–1058.

Vantrepotte, V., Loisel, H., Meriaux, X., Neukermans, G., Dessailly, D., Jamet, C.,Gensac, E., Gardel, A., 2011. Seasonal and inter-annual (2002–2010) variabilityof the suspended particulate matter as retrieved from satellite ocean colorsensor over the French Guiana coastal waters. Journal of Coastal Research,1750–1754.

Walsh, I.D., Gardner, W.D., 1992. A comparison of aggregate profiles with sedimenttrap fluxes. Deep-Sea Research Part a-Oceanographic Research Papers 39, 1817–1834.

Walsh, R.S., Hunter, K.A., 1992. Influence of phosphorus storage on the uptake ofcadmium by the Marine Alga Macrocystis-Pyrifera. Limnology andOceanography 37, 1361–1369.

Werdell, P.J. et al., 2013. Generalized ocean color inversion model for retrievingmarine inherent optical properties. Applied Optics 52, 2019–2037.

Wood, M.M., Particulate Matter in the South Atlantic Ocean. 1993. MS Thesis, TexasA&M University, 84 pp.